AI agents · SMBs Spain & Europe
Real AI implementation with Claude API + vector embeddings + n8n. Custom chatbots that answer from your catalogue and documentation, assistants for bookings or lead capture, automation of repetitive processes. No generic SaaS. No empty LinkedIn demos. Verifiable technical stack.
LinkedIn sold you AI as the answer to everything. You've watched gorgeous demos, tried a SaaS chatbot and discovered it can only answer what's already on the site. Meanwhile your team keeps replying to the same emails, fielding the same questions and running the same repetitive tasks. AI doesn't reach the business because between the hype and the technical reality nobody explains the gap.
Tidio, Intercom or some widget. They let you train it with 50 prefab sentences. When a customer asks anything slightly specific, the bot fires "let me transfer you to a human" and we're back to square one. Subscription paid, nothing changes.
Answering catalogue questions, qualifying form leads, translating product descriptions, drafting initial mailbox responses. Things your team does 50 times a day that a well-trained AI agent would handle in seconds with your brand voice.
Every agency sells "AI solutions" without showing a single line of code. 10x ROI promises, edited LinkedIn videos, €6,000 proposals where if you ask the stack they answer "GPT". No real stack means no real AI: just an OpenAI wrapper with markup.
You have a catalogue, order history, FAQs, internal manuals, technical documentation. All gold for an AI agent with vector embeddings: it answers with your real content, not invented responses. But nobody has shown you how to activate it.
Real technical implementation: Claude API or OpenAI per case, vector embeddings with your own documents, n8n for workflows and RAG (Retrieval-Augmented Generation) so the agent answers with your knowledge base, not generic invented replies. Integration with your existing site or deployment as a widget. No template SaaS, no closed-platform lock-in.
Ingestion of your documents (PDFs, URLs, spreadsheets, internal FAQs) into a vector database. When a customer asks, the agent searches the relevant chunk in your base and answers with that specific content, not generic replies. Personality and tone matched to your brand. Escalates to human with full context when needed. No "let me transfer you" as the default response.
Agent that collects lead data, qualifies buying intent, books a slot or call, routes to the right channel (WhatsApp, email, Cal.com) and drops context in your CRM. Adapted to your business funnel: restaurant → bookings, B2B service → lead scoring, academy → admission calls. We design the logic together in the brief.
If your customer searches "something comfortable for winter running" instead of "thermal running shoe", traditional search returns zero results. With vector embeddings the system understands intent and surfaces the relevant products. Applies to e-commerce, technical documentation repositories, internal manuals, legal bases. Integration with your existing stack (WooCommerce, Shopify if applicable, custom code).
Self-hosted n8n with Claude nodes and integrated tools. Workflows that classify incoming emails, draft initial responses in your tone, translate product descriptions into 3 languages, score leads using your business criteria, fire internal alerts when X condition is met. Repetitive work stops costing hours and stops failing on human fatigue.
You don't need a new site. The agent deploys as an embedded JS widget on your current site (WordPress, custom code, Shopify) or as an API your team consumes from anywhere (CRM, WhatsApp Business, internal Slack). If the site is very old, I warn you about technical limitations. If we're building it in parallel with Conversion Web, the widget ships built-in.
Why not generic SaaS or "AI agency"
Agency that wants to offer white-label AI agents to clients? See B2B agency partner. B2B rate with margin for your agency.
Technical stack
Main LLM: Claude API (Anthropic). Alternative OpenAI per case. Embeddings: OpenAI text-embedding-3 or BGE multilingual for peninsular languages. Vector base: Supabase pgvector, Pinecone or Qdrant by volume. Orchestration: n8n self-hosted or LangChain by complexity. Deployment: embedded JS widget or REST API on your hosting. Monitoring: logs and API costs tracked from day one so you know exactly what each conversation costs.
Weeks from brief to deployment
Ownership of the vector base
Generic SaaS in the pipeline
Direct technical point of contact
20-min call to understand what the agent solves (support, bookings, scoring, search) and what data you have (PDFs, CRM, catalogue, documentation). Written technical brief in 48 h with scope, proposed architecture and closed price range.
System design: chosen LLM (Claude/OpenAI), vector base, RAG layers, integrations with your systems. Knowledge base ingestion (chunking, embeddings, indexing). Your team validates content before training.
Agent implementation with tuned prompts, brand tone of voice, human escalation logic, channel integration (web, WhatsApp, email). Testing with real cases from your team. Iteration until responses are correct and sound like your business.
Deploy as widget on your site or API as scoped in the brief. Dashboard with conversations, API costs and usage metrics from day one. Optional monthly retainer to add documents, tune prompts and refine responses with real feedback.
Price by scope.
No surprise invoices.
Price depends on scope: data volume to index, agent complexity, integrations with your systems and deployment channels. I send a closed range after the 48 h technical brief. LLM API and vector base costs are separate, tracked in your own dashboard so you see exactly what each conversation consumes. No opaque invoices, no hidden usage pricing.
AI implementation
An AI agent shines when it has someone to attend. If the site doesn't capture visits or convert traffic that arrives, AI just automates silence. The complete Burguera system connects the three pieces so the agent works on real traffic and converts more.
Multi-page hand-coded site, Lighthouse 95+ and tracking ready. Needed for the AI agent to integrate on a stable base.
Full monthly SEO: keywords, content, link building. Brings qualified traffic to the agent, not cold leads.
AI agents with your knowledge base, assistants and automations. This is what you're already looking at.
One technical point of contact from brief to deploy.
No middlemen agencies, no lost handoffs.
What I get asked most before implementing an AI agent.
Closed project from €1,500 up to €8,000 by scope (basic FAQ chatbot vs complex assistant with CRM integrations). Setup + monthly retainer from €400/month. Price depends on three factors: data volume to index, integration complexity with existing systems and deployment channels (web, WhatsApp, CRM). LLM API costs (Claude/OpenAI) and vector base are separate and you pay them directly to the provider, tracked in your dashboard so you see exactly what each conversation consumes. Technical brief in 48 h after initial call · always free.
A template chatbot answers with what you type into a config box: 50-100 prefab sentences. If the customer asks anything unexpected, it fires "let me transfer you". An AI agent with vector embeddings indexes all your real documentation (catalogue, FAQs, manuals, history) and when a customer asks, the system finds the relevant chunk and answers with that specific information. It can also make decisions (book, schedule, qualify lead), escalate to human with full context, and learn from feedback. The technical difference is called RAG (Retrieval-Augmented Generation).
Between 2 and 6 weeks from signed brief. Simple projects (FAQ + lead capture on web): 2-3 weeks. Medium projects (chatbot + CRM + WhatsApp integration): 3-4 weeks. Complex projects (semantic search + n8n automations + multi-channel): 4-6 weeks. The timeline holds if your team delivers data and validates responses in agreed iterations. If the knowledge base is scattered or outdated, that extends the timeline and I tell you from the technical brief.
Three things: (1) business documentation in any format (PDFs, your website URLs, spreadsheets, catalogue exports, internal FAQs in Notion or Google Docs), (2) access to systems the agent must consult or update (CRM, WhatsApp Business, web form, booking calendar), (3) a reference person on your team to validate agent responses during 2-3 training iterations. If data is scattered or outdated, I help you structure it as part of the project.
No. The agent deploys as embedded JS widget on your existing site (WordPress, custom code, Shopify, Webflow), or as REST API your team consumes from anywhere (CRM, WhatsApp, Slack, internal panel). If the site is very old or has serious technical issues, I warn you from the brief. If we're building it in parallel with Conversion Web, the widget ships built-in without extra integration work.
Yes. The vector knowledge base updates with new documents without retraining from scratch: you add a PDF, a page or an export and it's indexed in minutes. Integrations with new systems (a new CRM, another messaging channel, an extra language) are added modularly. The monthly retainer model covers these periodic updates. If volume grows much (10x conversations/month), we review architecture (dedicated vector base, caching) without replacing the system.
Yes, and transparency here is part of the differentiator. (1) It doesn't replace human judgement on complex or sensitive decisions (legal, medical, serious financial): it should always escalate to human. (2) It can "hallucinate" (invent) if the knowledge base has inconsistencies or contradictions: that's why data curation is part of the process. (3) It needs maintenance when the catalogue, prices or business processes change: that's why the monthly retainer model exists. (4) It doesn't generate magical month-1 ROI: like SEO, AI is leverage, not a shortcut. If someone promises that, they're selling LinkedIn demos.
20 min, no commitment. You tell me what the agent solves for your business (support, bookings, scoring, search), what data you have and what channels you need. Written technical brief in 48 h with proposed architecture and closed price range. If it doesn't fit, I point you honestly elsewhere.
Book a 20-min call20-min technical call.
You and me, no stackless salespeople. You tell me the business problem to solve and what data you have.
What happens next.
Written technical brief in 48 h with scope, architecture, price range and timeline. You decide calmly.
